Abstract
Background: Autologous osteochondral transfer is an option for the
treatment of articular defects. However, there are concerns about graft
integration and the nature of the tissue forming the cartilage-cartilage
bridge. Chondrocyte viability at graft and recipient edges is thought to be an
important determinant of the quality of repair. The purpose of the present
study was to evaluate early cell viability at the edges of osteochondral
grafts from ex vivo human femoral condyles.
Methods: Fresh human tissue was obtained from eleven knees at the
time of total knee arthroplasty for the treatment of osteoarthritis.
Osteochondral cylinders were harvested with use of a 4.5-mm-diameter
mosaicplasty osteotome from regions of the anterolateral aspect of the femoral
condyle that were macroscopically nondegenerate and histologically
nonfibrillated. Plugs were assessed for marginal cell viability by means of
confocal laser scanning microscopy.
Results: The diameter of the cartilaginous portion of the
osteochondral plugs was a mean (and standard error of the mean) of 4.84
± 0.12 mm (as determined on the basis of three plugs). This value was
approximately 300 µm greater than the measured internal diameter of the
osteotome. There was a substantial margin of superficial zone cell death (mean
thickness, 382 ± 68.2 µm), with >99% cell viability seen more
centrally (as determined on the basis of five plugs). Demiplugs were created
by splitting the mosaicplasty explants with a fresh number-11 scalpel blade.
The margin of superficial zone cell death at the curved edge was significantly
greater than that at the site of the scalpel cut (390.3 ± 18.8 µm
compared with 34.8 ± 3.2 µm; p = 0.0286). Similar findings were
observed when the cartilage alone was breached and the bone was left intact,
with the margin of superficial zone cell death being significantly greater
than that obtained in association with the straight scalpel incision (268
± 38.9 µm compared with 41.3 ± 13.4 µm; p = 0.0286). The
margin of superficial zone cell death showed no increase during the
time-period between fifteen minutes and two hours after plug harvest. A
mathematical approximation of the mosaicplasty region suggested that early
cell death of this magnitude affects about one third of the superficial graft
area.
Conclusions: The results of the present study suggest that
mosaicplasty, while capable of transposing viable hyaline cartilage, is
associated with an extensive margin of cell death that is likely to compromise
lateral integration and articular reconstruction.
Clinical Relevance: The data suggest that there is a need to improve
the plug-harvest technique, which may improve graft-recipient healing and
clinical outcomes.
Osteochondral grafting is an attractive concept because the bone can
heal by creeping substitution and the articular cartilage can survive because
of adequate diffusional nutrition. Furthermore, cartilage is a privileged
immunological site for allografts because it is avascular and chondrocytes are
shielded by extracellular matrix. Unfortunately, articular cartilage has
minimal reparative
capacity1.
Over the past two decades, there have been encouraging clinical reports
concerning osteochondral
allografts2, lateral
patellar autologous grafting for the treatment of large defects at the
knee3, autologous
chondrocyte
implantation4,5,
and autologous osteochondral transfer
(mosaicplasty)6-8.
Mosaicplasty is an option for the treatment of smaller articular defects,
especially those due to trauma or osteochondritis
dissecans6-8.
The early results of one randomized, controlled trial favored autologous
chondrocyte implantation over autologous
transfer5. However,
the results of another trial (involving second-look biopsies) were more
equivocal, suggesting that fibrocartilage is the predominant repair tissue
following autologous chondrocyte
implantation8.
Hyaline cartilage has a complex avascular heterogeneous
structure9-11
in which chondrocytes constitute only 5% of the volume yet are solely
responsible for matrix maintenance. There are many data concerning
cartilaginous wounding and the healing
response9,11.
Partial-thickness defects do not heal, and full-thickness defects repair with
mechanically inferior
fibrocartilage1.
Chondrocyte viability is important in order for extracellular matrix to remain
healthy because, as noted by Gilmore and
Palfrey12, each
chondrocyte has the potential to maintain a discrete maximum volume of matrix
(a domain) surrounding it.
It has been well established that surgical maneuvers involve cell death at
the wound
edge13,14,
and chondrocyte death is ultimately associated with degeneration of the
surrounding matrix. Recent whole-animal experiments demonstrated a loss of
chondrocytes from the cartilage bordering the wound edge of partial-thickness
defects that had been generated with a custom-built planing
instrument15.
Trephine wounding of bovine articular cartilage explants induced cell death
over several
days16. On the
basis of ultrastructural and TUNEL (Tdt-mediated biotin-dUDP nick-end
labeling) analyses, it was suggested that cell death was at least partially
attributable to
apoptosis16.
Trephine wounding of immature bovine cartilage explants that had been
harvested off the bone was associated with a band of cell death.
Microautoradiography showed little isotope
incorporation14 in
this zone of death, suggesting that there was no new matrix synthesis or
proliferation in this zone. In addition, the mechanical impact associated with
osteochondral tamping into predrilled osseous holes causes superficial zone
cell death17.
Studies on the pathological effects of impact have shown cell death
colocalizing with articular surface
cracks18, mixed
necrotic and apoptotic
features19,
chondrocyte
apoptosis20, and
glycosaminoglycan
release20.
Hyaline cartilage is capable of surviving transfer. However, after
grafting, an absence of matrix-producing cells in the region of the
cartilage-cartilage interface makes integrative cartilage repair
unlikely13,14.
Seams of mechanically inferior tissue or gaps between grafts may affect joint
congruency and are likely to be a starting point for further
degeneration8.
Chondrocyte viability at the edge is likely to be a determinant of long-term
success. However, examination of the effect of osteochondral surgical
procedures on graft-edge viability, especially in human tissue, has been very
limited1,14.
The choice of method for the study of cell viability is crucial. In
contrast with conventional fixation, sectioning, and histological analysis,
confocal laser scanning microscopy allows three-dimensional visualization of
living cells in their native extracellular matrix and quantification of living
and dead cells. It also permits serial optical imaging and three-dimensional
reconstruction. The same volume of tissue can be imaged over time and from
more than one perspective. For example, cartilage blocks examined from a
superficial perspective subsequently can be examined transversely for
fibrillation. Confocal laser scanning microscopy therefore confers
considerable advantages over traditional microscopy.
The purpose of the present study was to use confocal laser scanning
microscopy in association with fluorescent indicators to determine early cell
viability at the human osteochondral graft edge.
Biochemicals
Biochemicals were obtained from Sigma (Poole, Dorset, United
Kingdom), unless otherwise stated. The culture medium was serum-free Dulbecco
modified Eagle medium (DMEM) with N-2-hydroxyethylpiperazine-N'-2-ethane
sulphonic acid (HEPES) at 25 mM. CellTracker Green (5-chloromethylfluorescein
diacetate; CMFDA) and propidium iodide (PI) were obtained from Molecular
Probes (Eugene, Oregon); stock solutions were propidium iodide (1-mM; aqueous)
and CMFDA (2.15-mM; dimethyl sulfoxide [DMSO]). Formaldehyde solution (10%
volume per volume in normal saline solution; pH 7.3) was obtained from Fisher
Scientific (Leicestershire, United Kingdom).
Ethical Approval
Ethical approval for the present study was obtained from the Fife Acute
Hospitals NHS Trust (Scotland) Ethics Committee. Additionally, informed
consent was obtained from patients undergoing total knee arthroplasty.
Human Osteochondral Explants
Resection specimens from patients undergoing total knee arthroplasty for
the treatment of osteoarthritis were placed immediately into DMEM and were
refrigerated overnight at 4°C (analogous to the temperature used for fresh
osteochondral
allografts21)
before use on the subsequent day. The lateral condylar area from the anterior
cut for total knee arthroplasty (Fig. 1,
A), analogous to the region used during
mosaicplasty21,22,
was used as a source for osteochondral plugs. A specimen was used only if
there was no macroscopic evidence of fibrillation, wear lines, or degeneration
in the relevant region. The rationale behind the use of this tissue is based
partly on evidence documenting the similarity between pristine human cartilage
and macroscopically nondegenerated cartilage specimens resected from
osteoarthritic
knees22. Of the
twenty-five knees examined, only eleven (from patients ranging from
fifty-three to seventy-nine years of age) had a lateral condyle that was
suitable for this study. Plugs were harvested with use of a sterile technique,
as utilized for tissue culture. A circular mosaicplasty osteotome with a
4.5-mm internal diameter (Fig. 1,
B) was used, according to the mosaicplasty
harvest technique, within twenty-four hours after the original operation. The
osteotome was applied perpendicular to the articular surface, and the cut was
deepened with use of a light (230-g) toffee hammer to harvest a bone-cartilage
composite plug measuring 7 to 9 mm in length. As during surgery, slight
toggling of the osteotome was used to allow dislodging. For experiments in
which tissue had to be preloaded with dye (see section entitled Dynamics
of Cell Death) or in which only the cartilage was breached (see section
entitled Margin of Cell Death When Bone Not Breached), an
osteochondral cuboid (Fig. 1,
C) was trimmed with use of a Stanley knife and
then was subjected to the plug-harvesting technique in a similar way. No
rotational movement was made during the harvesting procedure; all cuts
(including those performed with a number-11 scalpel blade with its straight
blade configuration) were made in push-through mode, with the direction of the
cutting force applied solely perpendicular to the blade edge (analogous to the
mode of cutting during mosaicplasty). The plug
(Fig. 1, D)
was removed from the inside of the circular chisel with use of an insert set
against the osseous surface. The articular surface was kept wet with DMEM
throughout the procedure as articular chondrocytes are known to be vulnerable
to the effects of
drying23. Care was
taken not to subject the hyaline surface to any abrasive or additional force
other than those associated with the harvesting procedure. Because of concerns
about blade sharpness, virgin tools were used when possible and no
mosaicplasty instrument was used to harvest more than six plugs
(Fig. 1, D).
This number is well within the operating limit of instrument use. The
cartilaginous portion of three plugs and the internal diameter of the
mosaicplasty chisel were measured with use of Vernier calipers (British
Aerospace, Bristol, United Kingdom) with an accuracy of ±10 µm.
Fluorescent Contrast Media
Fluorescent contrast media have been used previously to document cell
viability and death differentially in cartilage
explants11,14,24.
CMFDA is a membrane-permeant dye that is cleaved by intracellular esterases to
produce a fluorescent and membrane-impermeant product. The cytoplasm of viable
cells is marked as
green25.
Conversely, propidium iodide is a charged molecule that is capable of crossing
cell membranes only if they are damaged. Propidium iodide stains the nuclei of
dead cells red. Tissue was incubated with contrast medium concentrations that
were chosen to optimize imaging parameters (10.75 to 21.5 µM for CMFDA and
5 µM for propidium iodide).
Incubation and Fixation
For most experiments, plugs were incubated at 37°C for 120 minutes,
with exposure to propidium iodide (5 µM) and CMFDA (10.75 µM) for the
final thirty minutes. At the end of the incubation period, plugs were washed
in DMEM and were transferred to 10% formalin (volume per volume) in saline
solution. They were stored at 4°C and, after passage to 70% ethanol
(volume per volume), they were analyzed in water (at 21°C) on the
subsequent day. However, for the time-course experiments, explants were
exposed to propidium iodide (5 µM) and CMFDA (21.5 µM) in DMEM at
37°C for fifteen minutes before incisions were made with the mosaicplasty
osteotome and scalpel. They were then transferred for confocal laser scanning
microscopy and were maintained in DMEM, with contrast media at the same
concentration, at 21°C over the time-course indicated.
Microscopy
An upright Zeiss Axioskop LSM 510 confocal laser scanning microscope was
fitted with an objective (×10 dry) to acquire images of in situ
articular chondrocytes labeled with fluorescent dyes. A multitrack protocol
involving argon and HeNe (helium-neon) excitation allowed separation of the
fluorescence emitted from CMFDA (which stained the cytoplasm of live cells
green) and propidium iodide (which stained the nuclei of dead cells red).
CMFDA and propidium iodide were subjected to excitation wavelengths
(EX?) of 488 and 543 nm, respectively, with use of bandpass
and long pass filters to measure emission (500 to 550 nm and >560 nm,
respectively). Cartilage was viewed from both surface and transverse
perspectives, thereby allowing for the imaging of all zones in detail
throughout the full cartilage depth. Laser power, detector gain, and
sensitivity were adjusted to obtain optimal images without excessive bleaching
or saturation. Optical sections were obtained at approximately 10-µm
intervals over a depth of 170 µm.
Image Analysis
Images were overlaid and projected over appropriate depths with use of the
Zeiss Image Browser (Hertfordshire, United Kingdom). Mean margins of cell
death were calculated with use of measurements of the plug perimeter and the
associated sector area of cell death, according to the equation:
d=1000×r-[r2-(2.r.A/p)]1/2
where d is the margin of cell death (in µm), A is the measured
area (in mm2), r is the radius of the plug (in mm), and
p is the measured perimeter (in mm). For the 4.5-mm harvester, the
actual plug diameter was 4.84 mm (and the radius was 2.42 mm) (see Results
section). Values were recorded from multiple images for all tissues to ensure
imaging of the plug circumference in its entirety.
Tissue Grading
Anterolateral femoral condylar tissue was used only if there was no
macroscopic evidence of degeneration, fibrillation, or wear
lines24. The
samples were confirmed as being focally nondegenerate (grade 0 according to
the system described by Bush and
Hall24) by means of
microscopic transverse sectional imaging (confocal laser scanning microscopy;
×10 objective) to ensure that all zones were without fibrillation or
splitting on the basis of extracellular matrix autofluorescence and
tissue-staining.
Statistical Analysis
Generally, data are given as the mean and the standard error of the mean.
For measurements pertaining to the marginal zone of death, normality could not
be assumed on the basis of the small number of observations. Therefore, where
appropriate, nonparametric significance tests were used (GraphPad Prism;
GraphPad Software, San Diego, California).
Caliper Measurements
Initially, it was important to establish the precise dimensions of
the harvested plugs, both for subsequent analysis and to determine if the
fixation procedure itself caused any shrinkage or swelling artifact.
Osteochondral plugs were harvested from only macroscopically normal regions of
the anterolateral aspect of the femoral condyle. The diameters of the
cartilaginous portion of three plugs were measured with use of Vernier
calipers. The internal diameter of the mosaicplasty chisel measured 4.54 mm.
The plug diameter was greater, measuring 4.84 ± 0.12 mm (mean and
standard error of the mean) in DMEM and 4.84 ± 0.03 mm when fixed and
passed through alcohol to water. The diameter of the plugs was approximately
300 µm larger than the internal diameter of the circular osteotome.
Possible interpretations of these data are proposed in the Discussion
section.
Definition of the Marginal Zone of Cell Death
Analysis of five whole plugs (from three joints) showed a marginal zone of
cell death with a mean thickness of 382 ± 68.2 µm. Untraumatized
areas at the center of the plugs demonstrated >99% viability
(Fig. 2). There was
considerable intraplug heterogeneity with regard to the width of the marginal
zone of cell death, as shown in a range of three-dimensional summation views
over the most superficial 170 µm, the approximate penetrative limit of this
technique (Fig. 3). When viewed
from the side (again to a depth of approximately 170 µm), the curved edge
of four of the five plugs showed >99% cell death throughout the full depth
of cartilage, similar to the findings noted along the curved edge of the
demiplugs in later experiments.
Four plugs, each of which had been cut from a different joint, were used
for the demiplug experiments. In these experiments, the thickness of the zone
of cell death along the circular cut (the cut that had been made with use of
the mosaicplasty osteotome) was compared with that along a straight cut down
the center of the plug (the cut that had been made with use of a fresh
number-11 scalpel blade). From the superficial perspective, the thickness of
the zone of cell death along the curved cut (mosaicplasty osteotome) was
significantly greater than that along the straight cut (scalpel blade) (390
± 18.8 µm compared with 34.8 ± 3.2 µm; p = 0.0286,
Mann-Whitney test). A representative view that includes the junction of the
curved and straight edges is shown in
Figure 4, A.
Representative side views of the straight (scalpel blade) and curved
(osteotome) cuts are shown in Figure 4,
B and C, respectively. The osteotome cut
produced >99% cell death to the limits of laser penetration (approximately
170 µm) throughout the full depth of the cartilage.
Margin of Cell Death When Bone Not Breached
It was important to address the potential problem posed by the force
required to breach the subchondral bone in these samples as this force might
be greater than that required in younger joints without degenerative disease.
The mosaicplasty osteotome and number-11 scalpel blade incisions were made in
four osteochondral explants, each from a different joint, with care being
taken to penetrate the cartilage only (that is, not to transgress the
subchondral bone) before blade removal. These explants subsequently were
subjected to the same protocol as described in the Materials and Methods
section (incubation at 37°C for two hours, with exposure to contrast
medium for the last thirty minutes only) before fixation. The zone of cell
death at the curved edge was significantly greater than that at the straight
edge (268 ± 38.9 µm compared with 41.3 ± 13.4 µm; p =
0.0286, Mann-Whitney test). Figure
5 shows representative views from a single plug with hand-drawn
overlays as used for measurements. Figure
5, A and Figure 5,
B show the marginal zone of cell death at both the graft
and donor edges. Figure 5,
C shows the zone of cell death along the number-11
scalpel blade cut. Figure 5,
D shows the intersection between the straight (scalpel)
cut and the curved (osteotome) cut along with the associated zones of death.
Figure 5, E shows the
intersection of the two straight (scalpel) cuts in the center of the plug.
Dynamics of Cell Death
The dynamics of cell death were assessed over a two-hour period after
harvest. Explants were obtained with use of a Stanley knife and were
preincubated with propidium iodide and CMFDA for fifteen minutes before
harvesting (see Materials and Methods section). Demiplugs were harvested, as
described previously, to allow for a comparison of the thickness of the zone
of marginal cell death along the curved (osteotome) and straight (scalpel)
cuts. The graft edge zone of cell death showed no increase during the
time-period between fifteen minutes and two hours after the incision. The
results of three separate experiments are shown in
Figure 6, A, with
representative views from one experiment shown in
Figure 6, B (depicting
the zone along a curved cut made with use of a mosaicplasty osteotome) and
Figure 6, C (depicting
the zone along a straight cut made with use of a fresh scalpel blade). These
data indicate that from the earliest available time-point (fifteen minutes) to
two hours after harvest with the mosaicplasty osteotome, there was no
progression in the margin of cell death.
Pristine human tissues for experimentation are in short supply. The
bulk of the experimental work concerning cartilage wounding, integrative
cartilage repair, and osteochondral grafting has involved animal models, and
the results of such studies can only be extrapolated to the human situation
with caution. We have chosen an alternative approach involving the use of
macroscopically nondegenerate cartilage explants from joint surfaces harvested
from patients undergoing total knee replacement for the treatment of
osteoarthritis. The anterolateral aspect of the femoral condyle is often
spared macroscopically from the arthritic process and correlates with the
osteochondral harvest site for autografting (mosaicplasty). Macroscopically
nondegenerate cartilage from patients undergoing knee arthroplasty for
osteoarthritis previously has been found to be indistinguishable from normal
human cartilage in terms of basic biochemical, metabolic, and histological
parameters22.
Cell viability of this surface can be maintained at close to 100% if it is
passed rapidly to DMEM, maintained at 4°C, and used within twenty-four
hours. Nevertheless, there are reservations about using tissue (1) from joints
with established osteoarthritis, (2) from patients in this age-group, and (3)
after refrigeration overnight (even if use occurs within twenty-four hours).
Aging is known to affect cartilage and
chondrocytes9,26.
Hangody and Fules6
found that the results of mosaicplasty were better in younger as compared with
older patients and therefore proposed an upper age-limit of fifty years for
this procedure. In the present study, although there was occasional
chondrocyte clustering in some sections, the cartilage surface appeared normal
macroscopically and transverse views of the articular surface did not
demonstrate fibrillation or fissuring in the superficial zone (collagen
autofluorescence) when observed with confocal laser scanning microscopy
(Fig. 3, B and
C). It is not inconceivable that passage to DMEM and
storage at 4°C overnight might have altered the mechanical sensitivity of
chondrocytes. Nonetheless, there was a significant difference in the margin of
cell death associated with mosaicplasty and scalpel cuts to the same depth.
Furthermore, for fresh allograft procedures, storage at 4°C in Ringer's
lactate for up to forty-eight hours is accepted
practice21.
A further reservation is that excessive force might be required to harvest
plugs from this tissue. Subchondral sclerosis is a pathological feature of
osteoarthritis, and therefore the force required to harvest osteochondral
plugs from the joint resection specimens might be greater than that required
at the time of mosaicplasty in younger knees. Although osteochondral
harvesting was performed only on tissue that was grossly and histologically
nondegenerate, it was nevertheless important to perform experiments in which
only the cartilaginous tissue was breached and the plug was left in situ
(Fig. 5). The margin of cell
death was again significantly greater than that observed in association with a
single scalpel cut to the same level, supporting the contention that it is
trauma to the cartilage alone that is predominantly responsible for the
observed marginal zone of cell death. It is emphasized that number-11 scalpel
blade cuts constitute an internal negative control and that there is minimal
marginal cell death in relation to such incisions. With respect to the
thickness of the zone of marginal death, our findings contrast with those of
Williams et al.27,
who used a coring drill and found no evidence of "thermal damage"
at the periphery of 15-mm-diameter plugs when the periphery was compared with
the central portion of the graft.
Conversely, damage to the superficial zone of the graft may exceed that
indicated by early cell death in our model because further processes leading
to cell death may only become manifest after the end of our time-course (that
is, more than two hours after harvest). Also, during mosaicplasty, the graft
surface is subjected to direct trauma when it is impacted into a predrilled
hole. This tamping procedure is known to compromise chondrocyte
viability17.
Furthermore, the plugs used for our experiments were shorter (length, 7 to 9
mm) than those obtained during mosaicplasty (length, 15 to 25
mm)6-8,
which minimized the trauma due to frictional contact on the inner aspect of
the circular osteotome. The circumferential death indicates that the
mosaicplasty construct lacks superficial zone hyaline cartilage-cartilage
contact, suggesting a mechanically inferior construct in terms of spreading
contact stresses.
There are a variety of modes through which surgical procedures can damage
an articular surface. Cells near the cut surface may be exposed to a variety
of insults, including compression, tension, and shear forces. These forces
depend on parameters such as (1) the proximity of the blade (that is, the
proximity of the cell to the cut surface), (2) the type of cutting edge, (3)
the relation of the cutting edge to its supportive abrasive surface (the face
of the blade), and (4) the angulation of the cutting edge and surface to
planes in the extracellular matrix with which chondrocytes are associated. In
the present study, both number-11 scalpel blades and mosaicplasty osteotomes
were used solely in a push-through mode (i.e., with no sawing or rotatory
movements) in order to minimize abrasive and shear forces.
In response to trauma, cell death may occur rapidly (within minutes) as a
result of direct mechanical disruption, but it also may occur over a more
prolonged period. It is interesting that there was no progression in the
boundary of marginal cell death between fifteen minutes and two hours
(Fig. 6). There was an absence
of dye colocalization, suggesting leakage of the CMFDA product away from
cellular remains showing nuclear-binding of propidium iodide. This finding
supports our contention that punctate red-staining is indicative of cell death
rather than sublethal injury. Tew et
al.16 documented
cell death over a longer time-course at a trephined wound edge in both mature
and immature bovine cartilage explants; their data suggest cell death by a
combination of apoptosis and necrosis. Our documentation of early chondrocyte
death does not preclude other processes from occurring over a longer
time-period. Indeed, an understanding of the mechanisms underlying early and
late cell death may allow the development of cytoprotective strategies for the
graft edge.
Not surprisingly, there was intraplug heterogeneity with regard to the
width of the marginal zone of cell death
(Fig. 3), which may have been
due in part to the anisotropy of hyaline cartilage, as has been demonstrated
in compression28,
indentation
stiffness29, and
split-line studies of the femoral
condyle30. The
heterogeneity means that to derive an average thickness for the marginal zone
of death, it is important to have either a random sampling procedure or the
capability to image the entire plug or demiplug.
The approximate tenfold increase in thickness of the zone of cell death
when the cartilage was incised with use of the mosaicplasty osteotome as
opposed to the scalpel may have been related to a variety of features of the
surgical implement (or a combination of these features). These features
include the facts that (1) the cutting edge of the osteotome is squared-off,
(2) the taper to the cutting edge of the osteotome blade is less acute than
that of the scalpel blade, (3) the cutting edge of the osteotome is curved,
which may have deleterious consequences in terms of developing a plane of
cleavage through the matrix, and (4) the osteotome has an internal bevel to a
closed (circular) system and therefore the tissue at the plug margin is likely
to be compressed and subjected to increased abrasion or shear as it is passed
through the osteotome.
Measurements of the diameter of the cartilaginous portion of three plugs
with use of Vernier calipers suggested that there is considerable expansion of
the tissue after removal from the osteotome. It is possible that there may be
tissue compression against the edge of the osteotome due to the combination of
an internal bevel and a closed (circular) system. After removal from the
osteotome, the cartilage expands to its former area. Trauma to the wound edge
may cause subsequent swelling in DMEM as a result of osmotic flux into
unconstrained glycosaminoglycans. Furthermore, there may be internal stress
release following incision, causing tissue expansion.
Recently, Redman et
al.14 used immature
bovine cartilage explants that had been cut off the bone as a model cartilage
system with which to compare the effects of trauma caused by a blunt
instrument (a trephine) and that caused by a sharp instrument (a scalpel
blade). In response to blunt injury, the authors characterized a band of cell
death that extended approximately 100 µm from the wound edge.
The present study extends previous findings concerning excessive marginal
death at the graft edge because it uses the current mosaicplasty technique as
a directly relevant surgical paradigm. However, on the basis of our results,
we cannot yet designate a particular feature of osteotome design as the cause
of marginal cell death. Recently, Evans et
al.31 compared the
effects of two harvest procedures (manual punch and power-harvesting) on
chondrocyte viability and documented more damage in the power-trephine group.
We are engaged in additional research concerning blade geometry and harvesting
in order to identify improvements to minimize chondrocyte death at the graft
edge.
In conclusion, we observed a marked marginal zone of cell death (with a
thickness of just under 400 µm) in osteochondral plugs within two hours
after harvest with use of the mosaicplasty technique. Using a planar model for
close-packed circles (see Appendix), we estimated that approximately 24% of
the graft surface is nonviable because of marginal cell death, with an
additional 9% consisting of the gap between plugs. Therefore, approximately
one-third of the mosaicplasty surface is not comprised of viable hyaline
cartilage. This may be an underestimation given that Hangody et
al.7 estimated the
filling rate for a close-packed circle configuration to be 80% (as opposed to
our calculation of 91%). Our data suggest that there is scope for improvement
of mosaicplasty harvest and graft techniques. This may be reflected in better
long-term graft viability, superior integrative repair, and, as a result,
improved clinical outcomes.
A detailed description of the planar model is available with the electronic
versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?
Note: The authors thank Mr. I.J. Brenkel, Mr. R.E. Cook, Ms. J.
McEachan, and Mr. T.I.S. Brown (Fife Acute Hospitals) for providing the knee
resection specimens; Mr. J. Lissimore for his expert materials and engineering
advice and apparatus construction; and Smith and Nephew for kindly providing
the mosaicplasty harvest kits free of charge. This project was supported by
the Arthritis Research Campaign (H0621) and Wellcome Trust
(045925/Z/95/A).
Hunziker EB. Articular cartilage repair:
basic science and clinical progress. A review of the current status and
prospects. Osteoarthritis Cartilage.2002;10:
432-63.10432
2002
[PubMed][CrossRef]
Gross AE. Cartilage resurfacing: filling
defects. J Arthroplasty.2003;18(3 Suppl 1):
14-7.1814
2003
[PubMed][CrossRef]
Outerbridge HK, Outerbridge AR,
Outerbridge RE, Smith DE. The use of lateral patellar autologous grafts for
the repair of large osteochondral defects in the knee. Acta Orthop
Belg.1999;(65 Suppl 1):
129-35.129
1999
Richardson JB, Caterson B, Evans EH,
Ashton BA, Roberts S. Repair of human articular cartilage after implantation
of autologous chondrocytes. J Bone Joint Surg Br.1999;81:
1064-8.811064
1999
[PubMed][CrossRef]
Bentley G, Biant LC, Carrington RW,
Akmal M, Goldberg A, Williams AM, Skinner JA, Pringle J. A prospective,
randomised comparison of autologous chondrocyte implantation versus
mosaicplasty for osteochondral defects in the knee. J Bone Joint Surg
Br.2003;85:
223-30.85223
2003
[CrossRef]
Hangody L, Fules P. Autologous
osteochondral mosaicplasty for the treatment of full-thickness defects of
weight-bearing joints: ten years of experimental and clinical experience.
J Bone Joint Surg Am.2003;85(Suppl 2):
25-32.8525
2003
[PubMed]
Hangody L, Rathonyi GK, Duska Z,
Vasarhelyi G, Fules P, Modis L. Autologous osteochondral mosaicplasty.
Surgical technique. J Bone Joint Surg Am.2004;86(Suppl 1):
65-72.8665
2004
[PubMed]
Horas U, Pelinkovic D, Herr G, Aigner T,
Schnettler R. Autologous chondrocyte implantation and osteochondral cylinder
transplantation in cartilage repair of the knee joint. A prospective,
comparative trial. J Bone Joint Surg Am.2003;85:
185-92.85185
2003
[PubMed]
Stockwell RA. Biology of
cartilage cells. New York: Cambridge University Press;
1979.
1979
Buckwalter JA, Mankin HJ. Articular
cartilage. Part I: tissue design and chondrocyte-matrix interactions. J
Bone Joint Surg Am.1997;79:
600-11.79600
1997
Mainil-Varlet P, Monin D, Weiler C,
Grogan S, Schaffner T, Zuger B, Frenz M. Quantification of laser-induced
cartilage injury by confocal microscopy in an ex vivo model. J Bone
Joint Surg Am.2001;83:
566-71.83566
2001
Gilmore RS, Palfrey AJ. Chondrocyte
distribution in the articular cartilage of human femoral condyles. J
Anat.1988;157:
23-31.15723
1988
Bos PK, DeGroot J, Budde M, Verhaar JA,
van Osch GJ. Specific enzymatic treatment of bovine and human articular
cartilage: implications for integrative cartilage repair. Arthritis
Rheum.2002;46:
976-85.46976
2002
[CrossRef]
Redman SN, Dowthwaite GP, Thomson BM,
Archer CW. The cellular responses of articular cartilage to sharp and blunt
trauma. Osteoarthritis Cartilage.2004;12:
106-16.12106
2004
[PubMed][CrossRef]
Hunziker EB, Quinn TM. Surgical removal
of articular cartilage leads to loss of chondrocytes from cartilage bordering
the wound edge. J Bone Joint Surg Am.2003;85 Suppl 2:
85-92.8585
2003
[PubMed]
Tew SR, Kwan AP, Hann A, Thomson B,
Archer CW. The reactions of articular cartilage to experimental wounding: role
of apoptosis. Arthritis Rheum.2000;43:
215-25.43215
2000
[PubMed][CrossRef]
Nabavi-Tabrizi A, Turnbull A, Dao Q,
Appleyard R. Chondrocyte damage following osteochondral grafting using metal
and plastic punches: comparative study in an animal model. J Orthop
Surg (Hong Kong).2002;10:
170-2.10170
2002
Lewis JL, Deloria LB, Oyen-Tiesma M,
Thompson RC Jr, Ericson M, Oegama TR Jr. Cell death after cartilage impact
occurs around matrix cracks. J Orthop Res.2003;21:
881-7.21881
2003
[PubMed][CrossRef]
Chen CT, Burton-Wurster N, Borden C,
Hueffer K, Bloom SE, Lust G. Chondrocyte necrosis and apoptosis in impact
damaged articular cartilage. J Orthop Res.2001;19:
703-11.19703
2001
[PubMed][CrossRef]
D'Lima DD, Hashimoto S, Chen PC, Colwell
CW Jr, Lotz MK. Human chondrocyte apoptosis in response to mechanical injury.
Osteoarthritis Cartilage.2001;9:
712-9.9712
2001
[PubMed][CrossRef]
Aubin PP, Cheah HK, Davis AM, Gross AE.
Long-term followup of fresh femoral osteochondral allografts for posttraumatic
knee defects. Clin Orthop.2001;391 Suppl:
S318-27.391S318
2001
[PubMed][CrossRef]
Brocklehurst R, Bayliss MT, Maroudas A,
Coysh HL, Freeman MA, Revell PA, Ali SY. The composition of normal and
osteoarthritic articular cartilage from human knee joints. With special
reference to unicompartmental replacement and osteotomy of the knee. J
Bone Joint Surg Am.1984;66:
95-106.6695
1984
Mitchell N, Shepard N. The deleterious
effects of drying on articular cartilage. J Bone Joint Surg Am.1989;71:
89-95.7189
1989
[PubMed]
Bush PG, Hall AC. The volume and
morphology of chondrocytes within nondegenerate and degenerate human articular
cartilage. Osteoarthritis Cartilage.2003;11:
242-51.11242
2003
[PubMed][CrossRef]
Poole CA, Brookes NH, Clover GM.
Confocal imaging of the human keratocyte network using the vital dye
5-chloromethylfluorescein diacetate. Clin Experiment
Ophthalmol.2003;31:
147-54.31147
2003
[CrossRef]
Martin JA, Buckwalter JA. Aging,
articular cartilage chondrocyte senescence and osteoarthritis.
Biogerontology.2002;3:
257-64.3257
2002
[PubMed][CrossRef]
Williams SK, Amiel D, Ball ST, Allen RT,
Wong VW, Chen AC, Sah RL, Bugbee WD. Prolonged storage effects on the
articular cartilage of fresh human osteochondral allografts. J Bone
Joint Surg Am.2003;85:
2111-20.852111
2003
Jurvelin JS, Buschmann MD, Hunziker EB.
Mechanical anisotropy of the human knee articular cartilage in compression.
Proc Inst Mech Eng [H].2003;217:
215-9.217215
2003
[PubMed]
Korhonen RK, Wong M, Arokoski J,
Lindgren R, Helminen HJ, Hunziker EB, Jurvelin JS. Importance of the
superficial tissue layer for the indentation stiffness of articular cartilage.
Med Eng Phys.2002;24:
99-108.2499
2002
[PubMed][CrossRef]
Below S, Arnoczsky SP, Dodds J, Kooima
C, Walter N. The split-line pattern of the distal femur. A consideration in
the orientation of autologous cartilage grafts. J Arthrosc Rel
Res.2002;18:
613-7.18613
2002
Evans PJ, Miniaci A, Hurtig MB. Manual
punch versus power harvesting of osteochondral grafts.
Arthroscopy.2004;20:
306-10.20306
2004
[PubMed][CrossRef]